copt.fmin_SAGA

copt.fmin_SAGA(f, g, x0, step_size=-1, n_jobs=1, max_iter=100, tol=1e-06, verbose=False, callback=None, trace=False)[source]

Stochastic average gradient augmented (SAGA) algorithm.

The SAGA algorithm can solve optimization problems of the form

argmin_x f(x) + g(x)
Parameters:
  • g (f,) – loss functions. g can be none
  • x0 (ndarray) – Starting point
Return type:

OptimizeResult
Returns:

The optimization result represented as a scipy.optimize.OptimizeResult object. Important attributes are: x the solution array, success a Boolean flag indicating if the optimizer exited successfully and message which describes the cause of the termination. See scipy.optimize.OptimizeResult for a description of other attributes.
Return type:

opt

References

Defazio, Aaron, Francis Bach, and Simon Lacoste-Julien. “SAGA: A fast incremental gradient method with support for non-strongly convex composite objectives.” Advances in Neural Information Processing Systems. 2014.

Rémi Leblond, Fabian Pedregosa, Simon Lacoste-Julien. “ASAGA: Asynchronous parallel SAGA”. Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (AISTATS). 2017.